Feasibility of using decades-old archival tissues in molecular oncology/epidemiology

Archival tissues are a bountiful resource for various studies. Polymerase chain reaction permits the use of such tissues for molecular biological analyses of disease causation. However, a comprehensive study using a large number of decades-old samples (20 or more years) for molecular oncology/epidemiology has never been shown to be feasible. We have relied upon the unique tumor registry of atomic bomb survivors to show that such studies are possible using 275 hepatocellular carcinoma and 41 skin cancer cases. We used 23 relatively recent thyroid papillary carcinoma cases from persons living in the vicinity of the Chernobyl nuclear reactor accident for comparison. Degradation of DNA is severe in autopsy hepatocellular carcinoma samples but can be compensated for by decreasing the polymerase chain reaction product size. Increasing the amount of DNA that is used by a factor of 8 improved amplification efficiency from approximately 60 to 80%. Age of the samples was not as great a problem as was the source of procurement. The extracted DNA can be used for all types of assays that require polymerase chain reaction amplification, such as restriction fragment length polymorphism, single-strand conformation polymorphism, and direct sequencing.